Mitigating Noisy Inputs for Question Answering
Mitigating Noisy Inputs for Question Answering
复制标题
DOI：
10.21437/interspeech.2019-3154
复制
发表时间：
2019-08
期刊：
影响因子：
--
通讯作者：
Denis Peskov;Joe Barrow;Pedro Rodriguez;Graham Neubig;Jordan L. Boyd-Graber
中科院分区：
文献类型：
--
作者：
Denis Peskov;Joe Barrow;Pedro Rodriguez;Graham Neubig;Jordan L. Boyd-Graber
自然语言处理系统通常位于不可靠输入的下游：机器翻译、光学字符识别或语音识别。例如，虚拟助手只能在理解你的演讲后回答你的问题。我们调查和减轻噪声的影响，从自动语音识别系统上的两个事实的问答（QA）任务。经验表明，将置信度集成到模型中并强制解码未知单词可以提高下游神经QA系统的准确性。我们在超过50万个嘈杂句子的合成语料库上创建和训练模型，并在Quizbowl和Jeopardy的两个人类语料库上进行评估！竞争。
Natural language processing systems are often downstream of unreliable inputs: machine translation, optical character recognition, or speech recognition. For instance, virtual assistants can only answer your questions after understanding your speech. We investigate and mitigate the effects of noise from Automatic Speech Recognition systems on two factoid Question Answering (QA) tasks. Integrating confidences into the model and forced decoding of unknown words are empirically shown to improve the accuracy of downstream neural QA systems. We create and train models on a synthetic corpus of over 500,000 noisy sentences and evaluate on two human corpora from Quizbowl and Jeopardy! competitions.